Re: [PATCH][ATM] use rtnl_{lock,unlock} during device operations (take 2)

chas williams (chas@cmf.nrl.navy.mil)
Fri, 06 Jun 2003 19:37:29 -0400


In message <20030606215406.GE21217@gaz.sfgoth.com>,Mitchell Blank Jr writes:
>The really gross part is that it uses kernel-land pointers as "opaque" VC
>descriptors, so basically if atmsigd goes nuts in can easily stomp all
>over the kernel's memory. Way back when I added a CAP_SYS_RAWIO check to
>the ATMSIGD_CTRL ioctl so at least there's no security hole but it's still
>damn gross (no offense, Werner :-)

yes, it pretty awful. the problem is that this will take some changes
to user space stuff as well to be done right. so i was hoping to put
it off till 2.7 and hopefully get something that's smp stable for
2.6.

>I agree that fixing that old cruft would be a lot more productive than
>trying to shoehorn the ATM devices into the netdevice framework.

i would be willing to review patches. my concern is to have something
stable for 2.6. making atm more compliant with the netdevice framework
will make it easier to maintain in the future (or so i hope).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/